Systems and methods for evaluating hearing health

Abstract

Systems and methods are provided for evaluating hearing health of a given user. An input audio signal is transformed into the frequency domain and a first and second hearing profile are applied to the input audio sample. The first hearing profile represents a healthy hearing standard and the second hearing profile is the given user's hearing profile. Using the hearing profiles, first and second perceptually relevant information (PRI) values are generated for the input audio sample. The first and second PRI values are analyzed against each other to generate a PRI index value for the given user, where the PRI index value is a hearing health index value for the given user. The given user's hearing profile may additionally be applied to differently processed audio samples to evaluate the amount of perceptual rescue offered by various digital signal processing algorithms.

What is claimed is: 
     
       1. A method comprising:
 transforming an input audio sample into the frequency domain; 
 applying a first hearing profile to the input audio sample to generate a first perceptually relevant information (PRI) value for the input audio sample, wherein perceptually relevant information is calculated by calculating perceptual entropy; 
 applying a second hearing profile to the input audio sample to generate a second perceptually relevant information value for the input audio sample, wherein perceptually relevant information is calculated by calculating perceptual entropy; and 
 analyzing the first and second PRI values to generate a PRI index value for a given user. 
 
     
     
       2. The method of  claim 1 , further comprising:
 processing the input audio sample using a parameterized processing function to generate a processed audio sample; 
 applying the second hearing profile to the processed audio sample to calculate a third perceptually relevant information value for the processed audio sample; and 
 analyzing the first, second, and third PRI values to determine a perceptual rescue effected by the parameterized processing function. 
 
     
     
       3. The method of  claim 2 , wherein:
 the perceptual rescue effected by the parameterized processing function is optimized by determining at least one processing parameter by a sequential determination of subsets of the at least one processing parameter, each subset determined so as to optimize the given user's perceptually relevant information for the input audio signal. 
 
     
     
       4. The method of  claim 3 , wherein:
 the parameterized processing function operates on one or more subband signals of the input audio signal; and 
 the perceptual rescue effected by the parameterized processing function is optimized by:
 selecting a subset of the subbands so that a masking interaction between the selected subset of subbands is minimized; 
 determining at least one first processing parameter for the selected subbands; and 
 determining at least one second processing parameter for an unselected subband based on the first processing parameters of adjacent subbands. 
 
 
     
     
       5. The method of  claim 4 , wherein the at least one second processing parameter for an unselected subband is determined based on an interpolation of the first processing parameters of adjacent subbands. 
     
     
       6. The method of  claim 4 , wherein the at least one first processing parameter is determined sequentially on a subband by subband basis. 
     
     
       7. The method of  claim 3 , wherein the parameterized processing function operates on one or more subband signals of the input audio signal, the method further comprising:
 selecting a subset of adjacent subbands; 
 tying the corresponding values of the at least one processing parameter for each subband of the selected subset of adjacent subbands; and 
 performing a joint determination of the tied processing parameter values by minimizing the given user's perceptually relevant information for the selected subset of adjacent subbands. 
 
     
     
       8. The method of  claim 7 , further comprising:
 selecting a reduced subset of adjacent subbands from the selected subset of adjacent subbands; 
 tying the corresponding values of the at least one processing parameter for each selected subband of the reduced subset of subbands; 
 performing a joint determination of the tied parameter values by minimizing the given user's perceptually relevant information for the reduced subset of subbands; 
 repeating the previous steps until a first single subband is selected as the reduced subset of adjacent subbands; and 
 determining a first processing parameter of the first single subband. 
 
     
     
       9. The method of  claim 8 , further comprising:
 selecting a second subset of adjacent subbands; 
 successively reducing the selected second subset of adjacent subbands until a second single subband is selected; 
 determining a second processing parameter of the second single subband; and 
 performing a joint processing of the first processing parameter of the first single subband and the second processing parameter of the second single subband. 
 
     
     
       10. The method of  claim 9 , wherein the joint processing of the first and second processing parameter comprises at least one of:
 joint optimization of the processing parameters for the derived single subbands; 
 smoothing of the parameters for the derived single subbands; and 
 applying constraints on the deviation of corresponding values of the processing parameters for the derived single subbands. 
 
     
     
       11. The method of  claim 1 , further comprising measuring the perceptual rescue of a plurality of parameterized processing functions by:
 separately processing the input audio sample using the plurality of parameterized processing functions to generate a plurality of processed audio samples; 
 applying the second hearing profile to each processed audio sample to calculate a perceptually relevant information value for each parameterized processing function, wherein the second hearing profile is obtained from the given user and comprises masking threshold and hearing threshold information; and 
 for each parameterized processing function, analyzing the calculated perceptually relevant information value to generate a corresponding PRI index value uniquely associated with the given user and the parameterized processing function. 
 
     
     
       12. The method of  claim 1 , wherein:
 the first hearing profile is obtained from a healthy human listener and the second hearing profile is obtained from the given user; and 
 the first hearing profile and the second hearing profile comprise one or more of masking threshold information and hearing threshold information. 
 
     
     
       13. The method of  claim 12 , further comprising:
 determining one or more portions of the masking threshold information from masking threshold curves, the masking threshold curves corresponding to the first hearing profile and the second hearing profile; and 
 determining one or more portions of the hearing threshold information from audiogram data. 
 
     
     
       14. The method of  claim 12 , wherein:
 the first PRI value represents a healthy standard PRI value; 
 the second PRI value represents the given user's unique PRI value; and 
 generating the PRI index value for the given user comprises calculating a ratio value based on the first PRI value and the second PRI value. 
 
     
     
       15. The method of  claim 1 , wherein one or more of the first hearing profile and the second hearing profile are derived from at least one of a suprathreshold test, a psychophysical tuning curve, a masked threshold test, a threshold test and an audiogram. 
     
     
       16. The method of  claim 1 , in which the second hearing profile is the given user's hearing profile and is estimated from demographic information of the given user. 
     
     
       17. An audio processing device comprising:
 at least one processor; and 
 at least one memory storing instructions, which when executed causes the at least one processor to:
 transform an input audio sample into the frequency domain; 
 apply a first hearing profile to the input audio sample to generate a first perceptually relevant information (PRI) value for the input audio sample, wherein perpetually relevant information is calculated by calculating perceptual entropy; 
 apply a second hearing profile to the input audio sample to generate a second perceptually relevant information value for the input audio sample, wherein perceptually relevant information is calculated by calculating perceptual entropy; and 
 analyze the first and second PRI values to generate a PRI index value for a given user. 
 
 
     
     
       18. The audio processing device of  claim 17 , wherein the instructions further cause the at least one processor to:
 process the input audio sample using a parameterized processing function to generate a processed audio sample; 
 apply the second hearing profile to the processed audio sample to generate a third perceptually relevant information value for the processed audio sample; and 
 analyze the first, second, and third PRI values to determine a perceptual rescue effected by the parameterized processing function. 
 
     
     
       19. The audio processing device of  claim 17 , wherein the instructions further cause the at least one processor to measure the perceptual rescue of a plurality of parameterized processing functions by:
 separately processing the input audio sample using the plurality of parameterized processing functions to generate a plurality of processed audio samples; 
 applying the second hearing profile to each processed audio sample to calculate a perceptually relevant information value for each parameterized processing function, wherein the second hearing profile is obtained from the given user and comprises masking threshold and hearing threshold information; and 
 for each parameterized processing function, analyzing the calculated perceptually relevant information value to generate a corresponding PRI index value uniquely associated with the given user and the parameterized processing function.